INDEX

Explanations

mentions of walking, physical activity, or mobility

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

encies

-0.84

anyahu

-0.77

nces

-0.75

ENCY

-0.72

 Aires

-0.71

mble

-0.69

xus

-0.68

 suppressed

-0.67

tery

-0.66

 uranium

-0.65

POSITIVE LOGITS

ways

0.95

bike

0.94

through

0.89

walking

0.88

walk

0.87

path

0.87

way

0.84

 paths

0.81

 Bike

0.81

hoe

0.79

Activations Density 2.002%

mentions of walking, physical activity, or mobility

No Comments

No Known Activations